vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

chore: fix typo in llama.cpp

Browse source
This commit is contained in:
alwqx 2024-05-02 09:11:33 +08:00
parent b0d943de17
commit e95be2907d
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama.cpp
View file

@ -2359,7 +2359,7 @@ static bool llama_kv_cache_init(
cache.recurrent = model.arch == LLM_ARCH_MAMBA; cache.recurrent = model.arch == LLM_ARCH_MAMBA;
cache.v_trans = !cparams.flash_attn; cache.v_trans = !cparams.flash_attn;
// TODO: support mixed reccurent Transformer architectues // TODO: support mixed reccurent Transformer architectures
// NOTE: (!a || b) is a logical implication (a -> b) // NOTE: (!a || b) is a logical implication (a -> b)
GGML_ASSERT(!cache.recurrent || n_embd_k_gqa == hparams.n_embd_k_s()); GGML_ASSERT(!cache.recurrent || n_embd_k_gqa == hparams.n_embd_k_s());
GGML_ASSERT(!cache.recurrent || n_embd_v_gqa == hparams.n_embd_v_s()); GGML_ASSERT(!cache.recurrent || n_embd_v_gqa == hparams.n_embd_v_s());